Upload folder using huggingface_hub
Browse files- README.md +12 -0
- config.json +49 -0
- config.yaml +164 -0
- model.safetensors +3 -0
README.md
ADDED
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
library_name: lerobot
|
3 |
+
tags:
|
4 |
+
- act
|
5 |
+
- model_hub_mixin
|
6 |
+
- pytorch_model_hub_mixin
|
7 |
+
- robotics
|
8 |
+
---
|
9 |
+
|
10 |
+
This model has been pushed to the Hub using the [PytorchModelHubMixin](https://huggingface.co/docs/huggingface_hub/package_reference/mixins#huggingface_hub.PyTorchModelHubMixin) integration:
|
11 |
+
- Library: https://github.com/huggingface/lerobot
|
12 |
+
- Docs: [More Information Needed]
|
config.json
ADDED
@@ -0,0 +1,49 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"chunk_size": 20,
|
3 |
+
"dim_feedforward": 3200,
|
4 |
+
"dim_model": 512,
|
5 |
+
"dropout": 0.1,
|
6 |
+
"feedforward_activation": "relu",
|
7 |
+
"input_normalization_modes": {
|
8 |
+
"observation.images.agentview": "mean_std",
|
9 |
+
"observation.images.robot0_eye_in_hand": "mean_std",
|
10 |
+
"observation.state": "mean_std"
|
11 |
+
},
|
12 |
+
"input_shapes": {
|
13 |
+
"observation.images.agentview": [
|
14 |
+
3,
|
15 |
+
256,
|
16 |
+
256
|
17 |
+
],
|
18 |
+
"observation.images.robot0_eye_in_hand": [
|
19 |
+
3,
|
20 |
+
256,
|
21 |
+
256
|
22 |
+
],
|
23 |
+
"observation.state": [
|
24 |
+
9
|
25 |
+
]
|
26 |
+
},
|
27 |
+
"kl_weight": 10.0,
|
28 |
+
"latent_dim": 32,
|
29 |
+
"n_action_steps": 20,
|
30 |
+
"n_decoder_layers": 1,
|
31 |
+
"n_encoder_layers": 4,
|
32 |
+
"n_heads": 8,
|
33 |
+
"n_obs_steps": 1,
|
34 |
+
"n_vae_encoder_layers": 4,
|
35 |
+
"output_normalization_modes": {
|
36 |
+
"action": "mean_std"
|
37 |
+
},
|
38 |
+
"output_shapes": {
|
39 |
+
"action": [
|
40 |
+
7
|
41 |
+
]
|
42 |
+
},
|
43 |
+
"pre_norm": false,
|
44 |
+
"pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1",
|
45 |
+
"replace_final_stride_with_dilation": false,
|
46 |
+
"temporal_ensemble_coeff": null,
|
47 |
+
"use_vae": true,
|
48 |
+
"vision_backbone": "resnet18"
|
49 |
+
}
|
config.yaml
ADDED
@@ -0,0 +1,164 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
resume: false
|
2 |
+
device: cuda
|
3 |
+
use_amp: false
|
4 |
+
seed: 1000
|
5 |
+
dataset_repo_id: kywch/mimicgen_stack_d0
|
6 |
+
video_backend: pyav
|
7 |
+
training:
|
8 |
+
offline_steps: 100000
|
9 |
+
num_workers: 4
|
10 |
+
batch_size: 64
|
11 |
+
eval_freq: 10000
|
12 |
+
log_freq: 200
|
13 |
+
save_checkpoint: true
|
14 |
+
save_freq: 20000
|
15 |
+
online_steps: 0
|
16 |
+
online_rollout_n_episodes: 1
|
17 |
+
online_rollout_batch_size: 1
|
18 |
+
online_steps_between_rollouts: 1
|
19 |
+
online_sampling_ratio: 0.5
|
20 |
+
online_env_seed: null
|
21 |
+
online_buffer_capacity: null
|
22 |
+
online_buffer_seed_size: 0
|
23 |
+
do_online_rollout_async: false
|
24 |
+
image_transforms:
|
25 |
+
enable: false
|
26 |
+
max_num_transforms: 3
|
27 |
+
random_order: false
|
28 |
+
brightness:
|
29 |
+
weight: 1
|
30 |
+
min_max:
|
31 |
+
- 0.8
|
32 |
+
- 1.2
|
33 |
+
contrast:
|
34 |
+
weight: 1
|
35 |
+
min_max:
|
36 |
+
- 0.8
|
37 |
+
- 1.2
|
38 |
+
saturation:
|
39 |
+
weight: 1
|
40 |
+
min_max:
|
41 |
+
- 0.5
|
42 |
+
- 1.5
|
43 |
+
hue:
|
44 |
+
weight: 1
|
45 |
+
min_max:
|
46 |
+
- -0.05
|
47 |
+
- 0.05
|
48 |
+
sharpness:
|
49 |
+
weight: 1
|
50 |
+
min_max:
|
51 |
+
- 0.8
|
52 |
+
- 1.2
|
53 |
+
lr: 5.0e-05
|
54 |
+
lr_backbone: 1.0e-05
|
55 |
+
weight_decay: 0.0001
|
56 |
+
grad_clip_norm: 10
|
57 |
+
delta_timestamps:
|
58 |
+
action:
|
59 |
+
- 0.0
|
60 |
+
- 0.05
|
61 |
+
- 0.1
|
62 |
+
- 0.15
|
63 |
+
- 0.2
|
64 |
+
- 0.25
|
65 |
+
- 0.3
|
66 |
+
- 0.35
|
67 |
+
- 0.4
|
68 |
+
- 0.45
|
69 |
+
- 0.5
|
70 |
+
- 0.55
|
71 |
+
- 0.6
|
72 |
+
- 0.65
|
73 |
+
- 0.7
|
74 |
+
- 0.75
|
75 |
+
- 0.8
|
76 |
+
- 0.85
|
77 |
+
- 0.9
|
78 |
+
- 0.95
|
79 |
+
eval:
|
80 |
+
n_episodes: 20
|
81 |
+
batch_size: 20
|
82 |
+
use_async_envs: true
|
83 |
+
wandb:
|
84 |
+
enable: true
|
85 |
+
disable_artifact: true
|
86 |
+
project: lerobot
|
87 |
+
notes: ''
|
88 |
+
fps: 20
|
89 |
+
env:
|
90 |
+
name: mimicgen
|
91 |
+
task: Stack_D0
|
92 |
+
state_dim: 9
|
93 |
+
action_dim: 7
|
94 |
+
episode_length: 200
|
95 |
+
meta: stack_d0_env.json
|
96 |
+
image_keys:
|
97 |
+
- agentview
|
98 |
+
- robot0_eye_in_hand
|
99 |
+
state_keys:
|
100 |
+
- robot0_eef_pos
|
101 |
+
- robot0_eef_quat
|
102 |
+
- robot0_gripper_qpos
|
103 |
+
use_delta_action: false
|
104 |
+
use_highres_image_obs: true
|
105 |
+
override_dataset_stats:
|
106 |
+
observation.images.agentview:
|
107 |
+
mean:
|
108 |
+
- - - 0.485
|
109 |
+
- - - 0.456
|
110 |
+
- - - 0.406
|
111 |
+
std:
|
112 |
+
- - - 0.229
|
113 |
+
- - - 0.224
|
114 |
+
- - - 0.225
|
115 |
+
observation.images.robot0_eye_in_hand:
|
116 |
+
mean:
|
117 |
+
- - - 0.485
|
118 |
+
- - - 0.456
|
119 |
+
- - - 0.406
|
120 |
+
std:
|
121 |
+
- - - 0.229
|
122 |
+
- - - 0.224
|
123 |
+
- - - 0.225
|
124 |
+
policy:
|
125 |
+
name: act
|
126 |
+
n_obs_steps: 1
|
127 |
+
chunk_size: 20
|
128 |
+
n_action_steps: 20
|
129 |
+
input_shapes:
|
130 |
+
observation.images.agentview:
|
131 |
+
- 3
|
132 |
+
- 256
|
133 |
+
- 256
|
134 |
+
observation.images.robot0_eye_in_hand:
|
135 |
+
- 3
|
136 |
+
- 256
|
137 |
+
- 256
|
138 |
+
observation.state:
|
139 |
+
- ${env.state_dim}
|
140 |
+
output_shapes:
|
141 |
+
action:
|
142 |
+
- ${env.action_dim}
|
143 |
+
input_normalization_modes:
|
144 |
+
observation.images.agentview: mean_std
|
145 |
+
observation.images.robot0_eye_in_hand: mean_std
|
146 |
+
observation.state: mean_std
|
147 |
+
output_normalization_modes:
|
148 |
+
action: mean_std
|
149 |
+
vision_backbone: resnet18
|
150 |
+
pretrained_backbone_weights: ResNet18_Weights.IMAGENET1K_V1
|
151 |
+
replace_final_stride_with_dilation: false
|
152 |
+
pre_norm: false
|
153 |
+
dim_model: 512
|
154 |
+
n_heads: 8
|
155 |
+
dim_feedforward: 3200
|
156 |
+
feedforward_activation: relu
|
157 |
+
n_encoder_layers: 4
|
158 |
+
n_decoder_layers: 1
|
159 |
+
use_vae: true
|
160 |
+
latent_dim: 32
|
161 |
+
n_vae_encoder_layers: 4
|
162 |
+
temporal_ensemble_momentum: null
|
163 |
+
dropout: 0.1
|
164 |
+
kl_weight: 10.0
|
model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b5e63847da1f4d5b394c6b67b6238608e4131b69cf818f7f400c4a21b4b71298
|
3 |
+
size 206389844
|